A characterization of the weak topology in the unit ball of purely atomic \(L_1\) preduals (Q2147803)

A Banach space \(X\) is said to have a [weakly] stable unit ball \(B_X\) if finite convex combinations of relatively [weakly] open subsets of \(B_X\) are again [weakly] open. Among earlier important achievements known to the reviewer are that \(C(K)\)-spaces have a weakly stable unit ball if and only if \(K\) is scattered and that Banach spaces with a weakly stable unit ball have the strong diameter 2 property (SD2P), that is, every finite convex combination of slices of \(B_X\) has diameter 2. Moreover, it is known that rotund spaces and spaces with the 3.2-intersection property have stable unit balls. Here are the main results from this paper: \begin{itemize} \item [(i)] The \(L_1\)-preduals with a weakly stable unit ball are exactly the purely atomic ones (both real and complex case). \item [(ii)] Let \(X\) be a Banach space with both stable and weakly stable unit ball and \(K\) scattered compact. Then \(C_0(K,X)\) has a weakly stable unit ball. \item [(iii)] A Banach space has a stable unit ball if and only if it has the property that for every \(x,s_1,s_2\in B_X\), \(\lambda\in (0,1/2]\) so that \(x=(1-\lambda)s_1+\lambda s_2\), and every \(\varepsilon>0\), there exist \(\delta>0\) and continuous functions \[ v_i:B_X(x,\delta)\cap B_X\to B_X(s_i,\varepsilon)\cap B_X,\ i=1,2, \] such that \(y=(1-\lambda)v_1(y)+\lambda v_2(y)\) for every \(y\in B_X(x,\delta)\cap B_X\). \item [(iv)] Infinite dimensional Banach spaces with a weakly stable unit ball satisfy the following diameter 2 property called ASSD2P: If \(S_1,S_2,\dots, S_n\) are slices of \(B_X\), there exist \(h_1\in S_1\), \(h_2\in S_2,\dots, h_n\in S_n\) and \(y\in B_X\) with \(\|y\|=1\) such that \(y\pm h_i\in S_i\) for all \(1,2,\dots, n\). \item [(v)] For Banach spaces with a weakly stable unit ball, ext(\(B_X\)) is weakly closed and nowhere weakly dense in \(B_X\). \end{itemize} In addition to all these results, the paper contains an interesting introduction that puts things into perspective. Reviewer's remark: Concerning results on spaces with stable unit balls, the interested reader may want to consult [\textit{A.~Suárez-Granero}, Trans. Am. Math. Soc. 330, No.~2, 677--695 (1992; Zbl 0777.46014)].